Negative He 10830 Flare Ribbons and Non-thermal Electrons

From RHESSI Wiki

Jump to: navigation, search


Nugget
Number: 406
1st Author: Graham KERR
2nd Author:
Published: 12 April 2021
Next Nugget: Short time scales
Previous Nugget: Tracing the sources of gradual solar energetic particle events
List all



Contents

Introduction

While flares are typically characterized by intense broadband brightenings, there can sometimes also be dimmings. A specific well-known but observationally elusive example would be a "Black Light Flare" (Ref. [1]; Nugget No. 144). Plausibly one could have a brief transient dimming of the optical continuum preceding an ordinary "white-light flare". While this phenomenon has not been convincingly observed in solar flares, a phenomenon probably akin to it has recently been described: negative flare ribbons observed in absorption lines of helium atoms, as discovered originally by Harold Zirin.

Now it gets a little complicated. Solar helium absorption lines can result from transitions of the triplet state of helium ("orthohelium"). While flare observations of the He I lines at 10830 Â and 5876 Â (the Fraunhofer D3 line) are rare, there have since been several other reports of flare-induced dimmings, including the first observations in stellar flares (Ref. [2]). An excellent solar example (Ref. [3}) showed that a narrow edge of the flare ribbons underwent a period of He I 10830 dimming before the flare brightening started. This ribbon front also displayed unique profile shapes of chromospheric lines (Mg II and C II). Since these unique profiles and the He I 10830 dimming were confined to the ribbon front, it is clear that these are manifestations of energy deposition into the chromosphere from newly energized magnetic loops. Figure 1 shows an example from Ref. [3] where the negative contrast is between 10-20% and persists for several tens of seconds.

Figure 1: Observations of He I 10830 Â dimming at the narrow leading edge of a flare ribbon, adapted from Ref. [3]. Top left is the context image. Top right is the flare observed in He I 10830 at 19:11:59UT, where the arrow indicates the location where the negative ribbon front is observed. Bottom left shows the same flare but from 18:56:14UT where the boxes show the areas over which the lightcurves in the bottom right panel were constructed. Bottom right shows lightcurves of the flare (red) and quiet Sun (green).

In order to begin to develop diagnostics of that flare energy deposition we must identify the physical mechanism that produces the dimming, which we outline here. See Ref. [4] for the whole story.

Populating Orthohelium

The He I triplet requires temperatures in excess of T~20 kK to populate by thermal collisions. Much study of these lines, both observationally and theoretically, has shed light on the formation of orthohelium in the quiet Sun. Coronal EUV emission irradiates the upper chromosphere, photoionizing He I to produceHe II. Electrons recombine with He II, but the recombinations can occur through excited states. This results in some amount of helium existing as orthohelium, which can then be photoexcited by photospheric radiation. An absorption line thus forms. This is known as the photoionization-recombination mechanism (PRM).

It has been suggested that an enhanced form of the PRM can be at work during flares. The flare-heated and mass-loaded corona emits more strongly in the EUV, increasing the production of orthohelium via PRM and producing a deeper absorption line profile.

An alternative to this PRM process comes from direct non-thermal collisional ionization of He I to He II by the electron beam itself. Non-thermal electrons ionize He I, and electrons then recombine again through excited levels so that there is an overpopulation of orthohelium relative to the quiet Sun scenario. This is known as the collisional ionization recombination mechanism, or CRM, to be disinguished from PRM.

He I 10830 in Flare Simulations

To analyze the relative importance of these mechanisms (PRM and CRM), our work (Ref. [4]) builds upon prior modelling efforts to identify which plays the dominant role. To do this we performed a simple experiment. A grid of fifty non-thermal electron-beam simulations were produced using the well-known RADYN radiation hydrodynamic code, covering a range of electron beam low-energy cutoffs, and a range of flare strengths. RADYN has been introduced in several previous Nuggets (No. 274, No. 289, No. 315, for example). Our simulations included both PRM and CRM. They were then repeated but with the omission of CRM (that is, He non-thermal collisional ionisations were simply switched off).

He I 10830 dimming was only present in the simulations that included He non-thermal collisional ionization. Therefore, it is the non-thermal collisional ionization (CRM) of helium that leads to the observed dimming of He I 10830 at ribbon fronts. Figure 2 shows He I 10830 lightcurves with various beam properties. Panel (A) includes helium non-thermal collisional ionizations whereas panel (B) omits them. It is clear that while both experiments show emission during the flare, only those simulations including CRM exhibit a period of dimming at flare onset.

Figure 2: Lightcurves of He I 10830 in several flare simulations forward-modelled using the RADYN code. Panel (A) shows the flares with He non-thermal collisional ionisations included, and panel (B) shows the same flare but with He non-thermal collisional ionisations omitted. A negative contrast at flare onset is only observed in the former.

Discrepancy with observations

While the magnitude of the dimming is captured by the models, the duration is not. In the models we can only produce dimmings that last for a few seconds, but the observations of Ref. [3] showed dimmings persisting for several tens of seconds (to over 100 s). Identifying the source of this discrepancy is important.

While we plan to experiment with multi-threaded modelling and different pre-flare atmospheres we note that we were able to produce an example of a sustained dimming, but only if we injected a small flux of non-thermal electrons for a period of time before ramping the energy flux to more typical flare energies. Figure 3 shows this experiment. A beam with instantaneous energy flux F = 5x108 erg/cm2 s was injected for 30 s, with the energy flux ramping up to 5x1010 erg/cm2 s. This is shown as the red line. In this case, the low energy flux at flare onset was not strong enough to drive the chromospheric temperature to above 20 kK so that the He I 10830 Â line remained dimmed for 30 s, before going into emission when the beam flux was increased.

Figure 3: An example of a simulation with sustained dimming, adapted from Ref. [4]. The red dashed line shows the injected instantaneous energy flux, and the black line shows the He I 10380 lightcurve. The dimming persists until the energy of the electron beam is increased at t ~ 30s

Conclusions

Although model details remain to be explored, we think that these RADYN calculations point clearly to the need for non-thermal collisions to explain the He I dimming phenomonenon, and we predict that true continuum "black light flares" will also occur by collisional over-ionization, as proposed in Ref. [1], when observations have improved sufficiently.

References

[1] "Black and white flares?"

[2] "The CARMENES search for exoplanets around M dwarfs. Variability of the He I line at 10830 Â"

[3] "Ultra-narrow Negative Flare Front Observed in Helium-10830 Â Using the 1.6 m New Solar Telescope"

[4] "He I 10830 Â Dimming During Solar Flares, I: The Crucial Role of Non-Thermal Collisional Ionisations"

Personal tools
Namespaces
Variants
Actions
Navigation
Toolbox